benchang1110/

TaiVision-pretrain-1M-v1.0

	TaiVision-pretrain-1M Dataset Card






	
	Dataset details

Dataset type: TaiVision-pretrain-1M-v1.0 is a traditional Chinese Image Captioning dataset. This dataset is a concatenation of the two datasets: liuhaotian/LLaVA-Pretrain and liuhaotian/LLaVA-CC3M-Pretrain-595K translated into traditional Chinese by yentinglin/Llama-3-Taiwan-8B-Instruct. License: Must comply with license of CC-3M, BLIP (if you use their synthetic caption).

Primary intended… See the full description on the dataset page: https://huggingface.co/datasets/benchang1110/TaiVision-pretrain-1M-v1.0.